This K01 Career Development Award application proposes a multidisciplinary 4-year training program to provide the candidate, Dr. Deirdre K. Tobias, with the experience and resources necessary to launch a successful career as an independent nutrition and diabetes epidemiologist. The training plan, developed closely with primary mentor Dr. JoAnn Manson and co-mentor Dr. Frank B. Hu, will broaden Dr. Tobias' repertoire of research methodologies, including biomarker analysis, metabolomics, and systems epidemiology, and significantly strengthen her chronic disease epidemiology skill set. Dr. Tobias is an emerging and prolific epidemiologic researcher of nutritional and lifestyle determinants of gestational diabetes (GDM) and type 2 diabetes; however, additional training in important analytical tools is necessary to achieve her long-term goals of being an independent and successful researcher in population health. Additionally, these activities will give her extensive training in project management, which is essential for her future independent research endeavors. The Harvard School of Public Health is a leading research and academic institution within the large Boston-area medical community. This network will provide vast resources and opportunities for Dr. Tobias during this important developmental period, including cutting-edge seminar series, relevant advanced coursework, training in the responsible conduct of research, and scientific conferences. The proposed research project offers the candidate an outstanding opportunity to become proficient in the novel area of metabolomics research by evaluating the role of promising individual metabolites with diabetes risk among high risk women with prior GDM in the Nurses' Health Study II. Stored fasting plasma samples from the NHS II cohort will be used to measure a well-validated panel of water soluble and amino acid metabolites, including branched chain and aromatic amino acids, gut flora metabolites (e.g., choline, TMAO), glutamate, glutamine, and the recently cited ?-aminoisobutyric acid (BAIBA). From this, an efficient prospective nested case-control design is proposed to analyze 200 diabetes cases and 200 matched controls with a history of GDM to identify metabolites associated with incident type 2 diabetes. Additionally, the project includes investigation of the known dietary sources for several of these metabolites with type 2 diabetes risk, among the 6,400 women with GDM in NHS II. This cohort is uniquely well-suited to assess these relationships, due to repeated dietary and lifestyle measures and 25 years of follow-up. Findings from the research proposed in this application may inform public health strategies to prevent type 2 diabetes in a high risk population, while increasing our understanding of the mechanisms by which diet confers diabetes risk among women with a history of GDM. The outstanding training opportunities in novel research areas such as metabolomics and systems epidemiology with key leaders in the fields will greatly enhance the skills and capabilities of the candidate, and position her for a successful and productive career as a nutrition and diabetes epidemiologist.